Article

A layer-wise Hierarchical and Attentional Analysis of Argument Structure Constructions in BERT Using Naturalistic Corpora

2026-06-12

Abstract excerpt

<title>Abstract</title> <p>Understanding whether large language models develop human-like representations of abstract grammatical knowledge is a central question in computational linguistics and cognitive science. This study investigates how the Bidirectional Encoder Representations from Transformers (BERT) model processes four fundamental Argument Structure Constructions (ASCs), Resultative, Caused-Motion, Ditra...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 22f42d5d-b7ec-5503-8d7e-e4ed970df47d
DOI: 10.21203/rs.3.rs-9637584/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

A layer-wise Hierarchical and Attentional Analysis of Argument Structure Constructions in BERT Using Naturalistic CorporaDOI 10.21203/rs.3.rs-9637584/v1Read

Testing a computational model of causative overgeneralizations: Child judgment and production data from English, Hebrew, Hindi, Japanese and K’iche’Semantic proximity Read

Constructing a protolanguage: Reconstructing prehistoric languages in a usage-based construction grammar frameworkSemantic proximity Read

Form Composition: The Generative Structural Engine of Metaphors in Human CognitionSemantic proximity Read

The Emergent Symbolic Cognition Framework: Recursive Symbolic Generation as the Engine of General IntelligenceSemantic proximity Read

Speakers encode silent structures: Evidence from complementizer priming in EnglishSemantic proximity Read

Artificial Language LearningSemantic proximity Read

The roles of neural networks in language acquisitionSemantic proximity Read

Non-Human Models of Semantic CategorisationSemantic proximity Read

Actor Identification in Transitive Constructions in Mandarin ChineseSemantic proximity Read

Form Theory: The Conceptual Architecture of Human ThoughtSemantic proximity Read

Inductive biases and the geometry of concept learningSemantic proximity Read

Linguistic inputs must be syntactically parsable to fully engage the language networkSemantic proximity Read